Diastereoselective synthesis of some β-carboline derivatives from L-amino acids

Article abstract of DOI:10.1016/S0957-4166(02)00644-4

L-Ala and L-Val were used as chiral inductors in a series of reactions in which a Pictet-Spengler cyclization, completed under mild conditions was the key step. Diketopiperazines 7a and 8a having R configuration at the newly created stereogenic center wer

Full text of DOI:10.1016/S0957-4166(02)00644-4

TETRAHEDRON:
ASYMMETRY
Pergamon
Tetrahedron: Asymmetry 13 (2002) 2295–2297
Diastereoselective synthesis of some b-carboline derivatives from
-amino acids
L
Aleksandra Siwicka,a Krystyna Wojtasiewicz,a Andrzej Leniewski,a Jan K. Maurinb,c and
Zbigniew Czarnockia,*
aFaculty of Chemistry, Warsaw University, Pasteura 1, 02-093 Warsaw, Poland
bDrug Institute, Chełmska 30/34, 00-750 Warsaw, Poland
Institute of Atomic Energy, 05-400 Otwock-Swierk, Poland
c
´
Received 17 September 2002; accepted 8 October 2002
Abstract—
L
-Ala and
L
-Val were used as chiral inductors in a series of reactions in which a Pictet–Spengler cyclization, completed
under mild conditions was the key step. Diketopiperazines 7a and 8a having R configuration at the newly created stereogenic
center were obtained with good diastereoselectivity due to 1,4-chirality transfer. Surprisingly, -Pro promoted the formation of the
product with S configuration in the predominant diastereomers 9a and 10a. © 2002 Published by Elsevier Science Ltd.
L
1. Introduction
tions. Several approaches devoted specifically to the
stereocontrolled transformations of such compounds
have been developed1 and have proved successful when
compounds from the chiral pool were applied as start-
ing materials.2 Of the other possible starting materials,
amino acids seemed to be of fundamental importance
Alkaloids, especially those based on isoquinoline or
indole systems, are among the most important groups
of naturally occurring compounds as a result of their
biological activity and possible pharmaceutical applica-
Scheme 1.
* Corresponding author. Tel.: +48 22 822 02 11; fax: +48 22 822 59 96; e-mail: czarnoz@chem.uw.edu.pl
0957-4166/02/$ - see front matter © 2002 Published by Elsevier Science Ltd.
PII: S0957-4166(02)00644-4

2296
A. Siwicka et al. / Tetrahedron: Asymmetry 13 (2002) 2295–2297
due to their versatility and widespread use in chemical
synthesis. We would like to present herein our prelimi-
nary results on the stereoselective construction of indole
derivatives using amino acids as chiral inductors. The
idea for this work derives from a hypothesis, according
to which the peptide chain has an influence on the
stereochemistry in the biosynthesis of isoquinoline and
indole alkaloids.3
tion mixture. Initial attempts to apply the Pictet–Spen-
gler-type condensation with methanolic hydrogen
chloride solution resulted in low diastereoselectivities,
probably due to unfavorable hydrogen bonding situa-
tions. We decided, therefore, to change the solvent for
an aprotic one. Accordingly, we found that the best
diastereoselectivity was obtained when ethyl acetate
was applied as a solvent and the reaction was com-
pleted at room temperature.
The diastereomeric mixture of
-alanine derivatives 7a11
L
2. Results and discussion
and 7b had diastereomeric ratio (dr) of 91:9 while those
derived from
L
-valine 8a12 and 8b had dr of 93:7 (in
Recently, we have found that
phenylalanine and -proline could be used in the
L-alanine, L-valine L-
1
both cases the dr was established on the basis of H
NMR of the crude reaction mixture). X-Ray analysis of
the predominant diketopiperazine 8a13 confirmed our
expectations of the stereochemistry at the C-1 stereo-
genic centre (Fig. 1).
L
Pictet–Spengler condensation reaction for the diastereo-
selective construction of tetrahydroisoquinoline deriva-
tives under mild conditions.4 The work presented herein
examines the possibility of using
and -Val as chirality sources to allow the efficient
L
-Ala,
L-Pro
L
Following our previous findings in the synthesis of
isoquinolines,4 we expected a better stereochemical out-
come along with an opposite diastereomeric preference
and stereoselective construction of the b-carboline
skeleton.
when L-proline was used as the chirality source. Indeed,
The first step, as shown in Scheme 1, involved prepara-
tion of the amides 3 and 4,5 respectively, from N-
we were able to obtain diketopiperazine 916 as a result
of the Pictet–Spengler condensation performed under
mild conditions (HCl in ethyl acetate at room tempera-
ture). Unfortunately, despite exhaustive efforts, we were
unable to obtain a single crystal of 9a suitable for X-ray
crystallographic analysis. In the search for another
derivative, potentially more prone to form crystals suit-
blocked-N-methyl-L-amino acids in the presence of
BOP as a coupling mediator6 and subsequent deblock-
ing of the nitrogen atom under hydrogenolytic condi-
tions. Further BOP-mediated coupling of amides 37 and
48 with phenylpyruvic acid gave ketoamides 59 and 6,10
respectively, which were easily isolated from the reac-
Figure 1. The structure and ORTEP diagram for compound 8a.
Figure 2. The structure and ORTEP diagram for compound 10a.

A. Siwicka et al. / Tetrahedron: Asymmetry 13 (2002) 2295–2297
2297
able for X-ray diffraction analysis, we decided to apply
m-chlorophenylpyruvic acid as a starting material. The
required diketopiperazine 10a17 was finally obtained
with dr of 99:1. Subsequent X-ray analysis18 did indeed
reveal that the C-1 stereogenic centre had S configura-
tion (Fig. 2).
1.25–1.23 (3H, d, J=7.0 Hz); 13C NMR (CDCl3, 125
MHz): 174.8, 136.3, 127.3, 122.0, 121.9, 119.3, 118.7,
113.0, 111.2, 60.4, 39.14, 3.11, 25.4, 19.6.
8. Specific rotation for 4: [h]2D3 −19.2 (c 1.0, CHCl3).
9. Specific rotation for 5: [h]2D3 −77.2 (c 1.0, CHCl3).
10. Specific rotation for 6: [h]2D3 −52.3 (c 1.0, CHCl3).
11. Specific rotation for 7a: [h]2D3 −24.6 (c 1.0, CHCl3).
12. Selected data for 8a: mp 220–227oC, [h]D23 −53.2 (c 1.0,
CHCl3), 1H NMR (CDCl3, 500 MHz) l: 9.27 (1H, s),
7.50–7.05 (9H, m), 4.95–4.85 (1H, dd, J1=13.0 Hz, J2=
5.0 Hz), 3.81 and 3.85 (2H, ABq, J=14.5 Hz), 3.47, 3.25,
3.03, 2.73 (4H, m, -CH2-CH2-), 2.93 (3H, s), 1.16–1.02
(1H, m), 0.52–0.48 (3H, d, J=6.5 Hz), 0.82–0.78 (3H, d,
J=7.0 Hz), 13C NMR (CDCl3, 125 MHz): 166.6, 165.3,
136.0, 135.3, 132.7, 130.6, 130.3, 128.8, 128.6, 126.5,
122.5, 119.8, 119.5, 118.6, 111.5, 109.5, 68.2, 68.2, 66.0,
45.5, 45.4, 36.0, 33.3, 20.6, 18.3.
Interestingly, it therefore seems that the stereochemical
outcome of the Pictet–Spengler reaction is directly
dependent on the structure of the L-amino acid used.
Moreover, the results presented are relatively uncom-
mon examples of constructing the b-carboline skeleton
with the Pictet–Spengler condensation from tryptamine
under the 1,4-chiral influence of amino acids. In sum-
mary, the satisfactory chemical yields together with the
high stereoselectivity of the method presented suggest
that this approach might be an attractive starting point
for the synthesis of indole alkaloids.
13. All measurements for 8a crystal were done at T=100 K
on Kuma KM4CCD k-axis diffractometer using
graphite-monochromated Mo-Ka radiation (u=0.71073
,
A). 1204 frames were collected. The data were corrected
References
for Lorentz and polarization effects. No absorption cor-
rection was applied. Data reduction and analysis were
carried out with the Kuma Diffraction (Wrocław) pro-
grams. The structure was solved by direct methods14 and
1. (a) Wang, H.; Ganesan, A. J. Org. Chem. 2000, 65,
4685–4693; (b) Kozmin, S.; Iwama, T.; Huang, Y.;
Rawal, V. J. Am. Chem. Soc. 2002, 124, 4628–4641; (c)
Ma, C.; Liu, X.; Li, X.; Flippen-Anderson, J.; Yu, S.;
Cook, J. J. Org. Chem. 2001, 66, 4525–4542; (d) Bailey,
P. D.; Hollinshead, S. P.; McLay, N. R.; Giordano, F. J.
Chem. Soc., Perkin Trans. 1 1993, 451–458; (e) Bailey, P.
D.; Hollinshead, N. R.; Wood, S. D. J. Chem. Soc.,
Perkin Trans. 1 1993, 431–439.
2. (a) Coppola, G.; Schuster, H. Asymmetric Synthesis:
Construction of Chiral Molecules Using Amino Acids;
John Wiley & Sons: New York, 1987; (b) Hanessian, S.
Total Synthesis of Natural Products: The ‘Chiron
Approach; Pergamon: Oxford, 1983; (c) Scott, J. In
Asymmetric Synthesis; Morrison, J., Ed.; Academic: New
York, 1984; Vol. 4, Parts 1 and 2.
3. Krejcarek, G.; Dominy, B.; Lawton, R. Chem. Commun.
1968, 1450–1452.
4. Zawadzka, A.; Leniewski, A.; Maurin, J. K.; Woj-
tasiewicz, K.; Czarnocki, Z. Org. Lett. 2001, 3, 997–999.
5. All new compounds described here presented spectro-
scopic and elemental analysis to confirm their structures.
Selected details are provided below.
refined using SHELXL.15 The Fo >2s(Fo ) criterion was
used only for calculating R factors and is not relevant to
the choice of reflections for the refinement. C25H27N3O2,
M=401.50, monoclinic space group P21; a=14.083(3),
2
2
,
b=11.102(2), c=14.415(3) A, i=113.75(3)°, V=
3
3
,
2063.0(7) A , Z=4 and Dx=1.293 Mg/m . Colorless
crystal, v(Mo-Ka)=0.083 mm−1
, F(000)=856, 5377
reflections collected. Least squares on F2 (all reflections),
R=0.0514, wR2=0.0757 (observed).
14. Sheldrick, G. M. Acta Crystallogr. 1990, A46, 467–473.
15. Sheldrick, G. M. SHELXL-93. Program for the Refine-
ment of Crystal Structures, Univ. of Go¨ttingen, Ger-
many.
16. Specific rotation for 9a: [h]2D3 +68.6 (c 1.0, CHCl3).
17. Selected data for 10a: mp 205–208oC, [h]D23 +61.4 (c 1.0,
CHCl3).
18. X-Ray intensity data for 10a were measured at T=293 K
on a Kuma KM4 k-axis diffractometer with Mo-Ka
,
radiation (u=0.71073 A). 6278 unique reflections col-
lected. The data were corrected for Lorentz and polariza-
tion effects. No absorption corrections were applied. The
structures were solved by direct methods from SHELS14
and refined using SHELXL software.15 Crystal data for
compound 10a: C24H22ClN3O2, M=419.90, monoclinic
6. Castro’s reagent: Benzotriazol-1-yl-oxy-tris(dimethyl-
amino)phosphonium hexaflurophosphate, see: Castro, B.;
Evin, G.; Selve, C.; Seyer, R. Tetrahedron Lett. 1975,
1219–1222.
1
7. Selected data for 3: [h]2D3 −4.2 (c 1.0, CHCl3), H NMR
space group P21; a=9.0880(18), b=14.099(3), c=
3
,
,
16.127(3) A, i=92.20(3)°, V=2064.9(7) A , Z=4 and
Dx=1.347 Mg/m3. Colorless crystal, m(Mo-Ka)=0.211
mm−1, F(000)=876. Least squares on F2 (all reflections),
R=0.0436, wR2=0.1222 (observed).
(CDCl3, 500 MHz) l: 8.36 (1H, s, disappearing with
D2O), 7.7–7.0 (5H, m), 7.31–7.24 (1H, br.m, disappearing
with D2O), 3.7–3.52 (2H, br.m), 3.06–2.92 (2H, br.m),
3.06–2.92 (1H, br.m), 2.27 (1H, s), 1.44–1.28 (1H, br.s),